Week 1:
H.2 Study Design: Population-based studies
Bias = systematic error resulting in an estimated association between exposure
and outcome that deviates from the true association in a direction that depends
on the nature of the systematic error.
Recall bias and social desirability reporting bias are forms of measurement bias;
the systematic differences in recall and reporting between exposure or outcome
groups wo have dissimilar characteristics can lead to confounding of the results.
A cross-sectional survey is a type of observational or descriptive study. Cross-
sectional surveys are known as prevalence surveys, since they can be used to
estimate the prevalence of disease in a population.
Analysis of cross-sectional data: The data can be analysed using correlations
between exposures and outcomes.
Problems with cross-sectional studies: The exposure and disease or outcome are
measured at the same time, it is not possible to say which is cause and which is
effect.
Confounding in observational studies is the main reason why we perform
intervention studies if we want to prove that diet/ nutrients cause disease.
Confounding variables need to be identified, measured and controlled for in
analyses.
A confounder can provide an alternative explanation for an apparent association
between a dietary exposure and a disease/health outcome. For example a
positive association between increased coffee consumption and reduces birth
weight could be explained by higher smoking levels, because it is well
established that smoking during pregnancy is associated with increased coffee
consumption (figure 2.7).
,H3 Study Design: Intervention Studies
Causality (toevalligheid) cannot be demonstrated using a observational study. To
demonstrate cause and effect requires an intervention study in which
consumption of a nutrient, food or diet is altered in a controlled way and the
effect on selected outcomes is measured. Intervention studies are higher up the
hierarchy of scientific evidence than observational studies.
Randomised controlled trials: Parallel and cross-over
Without a control group, it is inappropriate to make cause-and-effect statements
about an intervention, as other factors may be responsible for the effects
observed. Having a control group means the ‘placebo effect’ can be assessed.
Parallel group studies: each participant receives only one of the nutrition
interventions (product A or B, or low intake or high intake). Figure 3.1
Cross-over studies: Participants receive all intervention under comparison and the
design specifies the order of interventions. This has the advantage that
comparisons between interventions van be made on a within-participant basis,
with a consequent improvement in the precision of comparisons and therefore in
the power of the study, and a reduction in the required sample size. Figure 3.1
RCT = randomised controlled trial
,Other less commonly used types of RCT include the factorial design (in which all
possible combination of two or more interventions are tested, therefore
permitting the evaluation of intervention interactions) and cluster randomised
design (in which the unit of randomisation is not the individual but a cluster of
individuals defined, for example, by family, school class or primary care group).
Analytical variability:
Laboratory analytical methods should be precise, accurate, sensitive and specific,
and these performance characteristics should be recorded in a file of standard
operating procedures (SOPs). Intra-laboratory analytical variability should be
minimised by using automated equipment to analyse samples in duplicate or
triplicate, in batches that represent the range of interventions, participants and
sampling times, with suitable internal and external standards and participation in
quality assurance programmes.
Biomarkers that have high methodological variability will often require a larger
number of trial participants to give the study adequate power.
Biological viability:
Biological viability arises from many factors (e.g. genetic background, circadian
rhythm, seasonal differences, menstrual cycle) and may introduce systematic
bias.
Size of study (power calculation)
It is essential to estimate the number of participants required for the study. A
study too small will likely fail to detect important differences between
interventions, while one that is too large may needlessly waste resources and
would be unethical.
The usual methods for sample size estimation require specification of the
magnitude of the smallest meaningful difference in the outcome variable. The
study must be sufficiently large tot have acceptable power to detect this
difference as statistically significant, and must take into account possible non-
compliance and the anticipated drop-out rate.
Methods to measure compliance
- Consumption under supervision
- Dietary records, such as food diaries or diet recall methods. But such self-
reported intake data are predisposed to errors. Thus the assessment of
tissue biomarkers as independent and objective measures of compliance is
preferred when possible (serum selenium or fatty acid)
, Week 2 – Body Composition
Anthropometry = literally meaning ‘’measurement of
humans’’
Skinfold measurement (huidplooi meting): Based on age and
gender you estimate body density.
How accurate is BMI?
BMI is not a very good test to test body composition.
two men who have the same BMI, but
have a different body fat percentage.
Body composition = ‘’breaking down the body into its core components’’
The estimation of body composition is an important component of health and
fitness assessments in individuals and populations. In particular, the amount
and distribution of fat and lean mass are important for health throughout the
lifespan, but also bone mineral and fluid spaces – as well as its distribution
Multi component model is the best method to assess body composition, but it is
unhandy to do.
Body composition
terminology
H.2 Study Design: Population-based studies
Bias = systematic error resulting in an estimated association between exposure
and outcome that deviates from the true association in a direction that depends
on the nature of the systematic error.
Recall bias and social desirability reporting bias are forms of measurement bias;
the systematic differences in recall and reporting between exposure or outcome
groups wo have dissimilar characteristics can lead to confounding of the results.
A cross-sectional survey is a type of observational or descriptive study. Cross-
sectional surveys are known as prevalence surveys, since they can be used to
estimate the prevalence of disease in a population.
Analysis of cross-sectional data: The data can be analysed using correlations
between exposures and outcomes.
Problems with cross-sectional studies: The exposure and disease or outcome are
measured at the same time, it is not possible to say which is cause and which is
effect.
Confounding in observational studies is the main reason why we perform
intervention studies if we want to prove that diet/ nutrients cause disease.
Confounding variables need to be identified, measured and controlled for in
analyses.
A confounder can provide an alternative explanation for an apparent association
between a dietary exposure and a disease/health outcome. For example a
positive association between increased coffee consumption and reduces birth
weight could be explained by higher smoking levels, because it is well
established that smoking during pregnancy is associated with increased coffee
consumption (figure 2.7).
,H3 Study Design: Intervention Studies
Causality (toevalligheid) cannot be demonstrated using a observational study. To
demonstrate cause and effect requires an intervention study in which
consumption of a nutrient, food or diet is altered in a controlled way and the
effect on selected outcomes is measured. Intervention studies are higher up the
hierarchy of scientific evidence than observational studies.
Randomised controlled trials: Parallel and cross-over
Without a control group, it is inappropriate to make cause-and-effect statements
about an intervention, as other factors may be responsible for the effects
observed. Having a control group means the ‘placebo effect’ can be assessed.
Parallel group studies: each participant receives only one of the nutrition
interventions (product A or B, or low intake or high intake). Figure 3.1
Cross-over studies: Participants receive all intervention under comparison and the
design specifies the order of interventions. This has the advantage that
comparisons between interventions van be made on a within-participant basis,
with a consequent improvement in the precision of comparisons and therefore in
the power of the study, and a reduction in the required sample size. Figure 3.1
RCT = randomised controlled trial
,Other less commonly used types of RCT include the factorial design (in which all
possible combination of two or more interventions are tested, therefore
permitting the evaluation of intervention interactions) and cluster randomised
design (in which the unit of randomisation is not the individual but a cluster of
individuals defined, for example, by family, school class or primary care group).
Analytical variability:
Laboratory analytical methods should be precise, accurate, sensitive and specific,
and these performance characteristics should be recorded in a file of standard
operating procedures (SOPs). Intra-laboratory analytical variability should be
minimised by using automated equipment to analyse samples in duplicate or
triplicate, in batches that represent the range of interventions, participants and
sampling times, with suitable internal and external standards and participation in
quality assurance programmes.
Biomarkers that have high methodological variability will often require a larger
number of trial participants to give the study adequate power.
Biological viability:
Biological viability arises from many factors (e.g. genetic background, circadian
rhythm, seasonal differences, menstrual cycle) and may introduce systematic
bias.
Size of study (power calculation)
It is essential to estimate the number of participants required for the study. A
study too small will likely fail to detect important differences between
interventions, while one that is too large may needlessly waste resources and
would be unethical.
The usual methods for sample size estimation require specification of the
magnitude of the smallest meaningful difference in the outcome variable. The
study must be sufficiently large tot have acceptable power to detect this
difference as statistically significant, and must take into account possible non-
compliance and the anticipated drop-out rate.
Methods to measure compliance
- Consumption under supervision
- Dietary records, such as food diaries or diet recall methods. But such self-
reported intake data are predisposed to errors. Thus the assessment of
tissue biomarkers as independent and objective measures of compliance is
preferred when possible (serum selenium or fatty acid)
, Week 2 – Body Composition
Anthropometry = literally meaning ‘’measurement of
humans’’
Skinfold measurement (huidplooi meting): Based on age and
gender you estimate body density.
How accurate is BMI?
BMI is not a very good test to test body composition.
two men who have the same BMI, but
have a different body fat percentage.
Body composition = ‘’breaking down the body into its core components’’
The estimation of body composition is an important component of health and
fitness assessments in individuals and populations. In particular, the amount
and distribution of fat and lean mass are important for health throughout the
lifespan, but also bone mineral and fluid spaces – as well as its distribution
Multi component model is the best method to assess body composition, but it is
unhandy to do.
Body composition
terminology
